Claim Fraud Indicator Screening Agent

Fraud screening should focus investigators' time where it is needed, without delaying honest claimants. For each new claim or update, this agent checks known indicators: a policy started shortly before the loss, repeated claims, mismatched dates, edited photos or links to other suspicious claims. It scores the claim and lists each indicator found. It then verifies each indicator with source data to remove false matches, such as two different people sharing a common name. If the verified score drops below the threshold, the claim is cleared for normal handling. If it stays above, the agent prepares a referral with the evidence. A low-value claim with one weak indicator is not referred. The special investigation team decides any referral and any action. Reasons are always listed. Edge case: photo metadata missing because of a messaging app is not counted as an indicator.